Betting on NFL win totals is a great way to invest for or against a team's upcoming season.
There are many factors that I look into when deciding on taking a side for each team's win total. The main data point I use to decide is the percent chance a team will go over or under any given number based on my season-long simulator that's based on my team ratings.
I use many metrics when coming up with my team ratings, but they factor in the team's current roster, any potential injuries/suspensions that would impact their given odds on any given week, and the coaching staff. I also attempt to dig into other metrics that indicate which teams were lucky/unlucky last season, which could result in the market being too high/low on a given team. Some of the metrics you will see me refer to include:
One-score record: A team's record in one-score games. The idea here is that we can assume teams should finish right around a 50/50 record in one-score games. If a team ends up going a few games above/below .500 in one-score games, it usually causes the market to over/under estimate that team based on their one-score luck. My logic is to always assume that their luck will regress back towards average the following season.